RICE Scoring

tool · management · organizing-schema

Prioritizes features by Reach, Impact, Confidence, and Effort.

RICE Scoring is a prioritization framework developed by Sean McBride at Intercom and published on the company's blog in 2017, designed to bring more quantitative discipline to product feature prioritization than purely qualitative methods like MoSCoW. The score for each candidate is computed as Reach × Impact × Confidence ÷ Effort, where Reach is the number of users affected per time period, Impact is the per-user effect on the relevant goal (typically scored 0.25 for minimal up to 3 for massive), Confidence is a percentage discount on Reach and Impact estimates reflecting estimation certainty, and Effort is the person-months of work required. Higher scores represent better impact-per-effort opportunities. The framework forces explicit quantification of what would otherwise be intuitive judgments, surfaces low-confidence inputs that warrant validation work, and supports comparison across heterogeneous initiatives. It's widely used in product management, particularly in B2C SaaS contexts where Reach is meaningfully measurable.

Originators

Sean McBride (Intercom) high

Year / Decade

2017 (Intercom blog publication) high

Primary sources

McBride, S. (2017). 'RICE: Simple prioritization for product managers', Intercom blog high

Core components

Reach (users affected per time period)
Impact (per-user effect on goal, typically 0.25 to 3 scale)
Confidence (percentage discount for estimation uncertainty)
Effort (person-months)
Score = Reach × Impact × Confidence / Effort
Comparative ranking of candidates
Originally developed for product feature prioritization

Primary use case

Product feature prioritization in product management; structured quantitative comparison of heterogeneous backlog items; basis for product-roadmap discussions; complement to qualitative prioritization frameworks.

Common criticisms

Quantification creates false precision — Impact and Confidence scores are largely guesses dressed in numbers
Reach is hard to measure for genuinely new features
Effort estimates are notoriously inaccurate, particularly for novel work
multiplicative formula amplifies small estimation errors
doesn't account for strategic value, dependencies, or option value of learning from experiments
can shift attention to high-Reach incremental features over higher-impact strategic bets (the local maximum trap)
confidence percentages often reflect optimism bias rather than genuine uncertainty assessment
works less well in B2B contexts where customer counts are small.

Lineage

Siblings: ICE Scoring, MoSCoW, Kano Model